Financial Risk Manager Part 1
Get started today
Ultimate access to all questions.
Which of the following best describes the Monte Carlo method for reinforcement learning?
Exam-Like
Community
TTanishq
A
It looks only one decision ahead when updating strategies.
B
It requires a lot of computational resources to simulate the entire environment.
C
The agent initially knows only the possible states and actions that it can take in an environment.
D
It requires a priori knowledge of the environment in order to learn a policy.
Explanation:
Explanation
Monte Carlo simulation involves simulating the entire environment to completion in order to generate sufficient samples for estimating the strategy. This makes it computationally intensive.
Why other options are incorrect:
-
A: Incorrect - this applies to temporal difference learning, not Monte Carlo method. In Monte Carlo, the strategy is updated based on the total future rewards obtained through interaction with the environment rather than just looking at one decision ahead.
-
C: Incorrect - this applies to temporal difference learning. In the Monte Carlo method, the agent has to wait until the completion of an episode to update the strategy.
-
D: Incorrect - Monte Carlo method does not require any a priori knowledge of the environment and can learn from the outcomes of actions taken in the environment rather than by explicitly learning a policy.
Comments
Loading comments...